INDEX

Explanations

references to specific decades in time, particularly the seventh decade of the 20th century and beyond

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.16

arma

-0.15

æł·çļĦ

-0.15

à¥įà¤°à¤¸

-0.14

ÑĥÐ¶Ðµ

-0.14

fred

-0.14

 then

-0.14

nt

-0.14

amus

-0.14

POSITIVE LOGITS

anism

0.16

-long

0.16

hower

0.16

YNAM

0.14

alley

0.14

asmus

0.14

 Smarty

0.13

.Deep

0.13

/*/

0.13

ç¸

0.13

Activations Density 0.025%

references to specific decades in time, particularly the seventh decade of the 20th century and beyond

No Comments

No Known Activations